concept Vector column type in category pyspark

appears as: Vector column type
Data Analysis with Python and PySpark

This is an excerpt from Manning's book Data Analysis with Python and PySpark.

For computing correlation between variables, PySpark provides the Correlation object. Correlation has a single method, corr, that computes the correlation between features in a Vector. Vectors are like PySpark arrays but with a special representation optimized for ML work (see the section called “Assembling the final data set with the Vector column type” for a more detailed introduction). In listing 10.17, I use the VectorAssembler transformer on the food data frame to create a new column continuous_features that contains a Vector of all our continuous features.